Big data/Machine Learning/AI as a policy savior

The Post has an article about how Virginia Congressman Don Beyer is getting a masters degree in machine learning so he can apply AI to government and creating better policy ("A 72-year-old congressman goes back to school, pursuing a degree in AI").

Over the years I've written about the hype about big data and "apps," making the point that there is plenty of "little data" and academic analysis of policy that is already being ignored, despite the richness it offers for improvement in policy and practice.

-- "All the talk of e-government, digital government, and open source government is really about employing the design method," 2012
-- "Creating the right program vs. the hype of big data," 2013
-- "Does the focus on big data mean we miss the opportunity for better use of "little data"," 2015
-- "For a lot of "urban problems" the issue isn't knowledge about what to do, but willingness to engage that knowledge," 2017

In fact, the Post has another current article, "How Black activists in Northern Virginia transformed the way children learn to read," about how despite persistent--over decades--lagging outcomes in Fairfax County Schools for black students in learning how to read the school district wasn't doing anything about it, until the local NAACP chapter stepped in.

Positive deviance theory: Hope amidst failure.  Years ago I read an article, "Your Company's Secret Change Agents, in the Harvard Business Review about "positive deviance" theory, how in even the worst performing organization there are pockets of excellence, and you can use those clusters of best practice to reshape the organization.

I believed in the approach for a long time, but the reality is that most organizations, even poorly performing ones, are highly resistant to change, improvement, and innovation, and they can counteract pockets of excellence with incredible resistance and inertia.

For example, I was really troubled by an example in Boston, where a junior high school improved to the point where it was an international best practice example, yet the people who helped the school succeed were pushed out, and over time the school reverted back to the mean of mediocre performance.  In fact, now it's slated for closure.

-- "A positive deviance failure in Boston: Timilty Middle School," 2022

What an incredible waste.

Like my frustration with DC Government.  After two terms, eight years, the reelected mayor is looking for innovative ideas?

-- "I can't help but laugh... DC Mayor's third term as a platform for transformation"

Similarly, the reelected Mayor of Toronto says he's best placed to reverse the city's last 8 years of decline, decline that he ruled over.

-- "Brutal performance art criticism of Toronto's Mayor, John Tory, and his "austerity" agenda"

Transformational Projects Action Planning. ("A wrinkle in thinking about the Transformational Projects Action Planning approach: Great public buildings aren't just about design, but what they do"). By contrast, I propose the concept of transformational projects action planning, leveraging master plans and projects to push innovation forward in a variety of ways.

TPAPs should be implemented at multiple scales: 

(1) neighborhood/district/city/county wide as part of a master plan; 

(2) within functional elements of a master plan such as transportation, housing, or economic development; and 

(3) within a specific project (e.g., how do we make this particular library or transit station or park or neighborhood "great"?); in terms of both 

(4) architecture and design; and 

(5) program/plan for what the functions within the building accomplish.

At the master plan scale, I've written about this in terms of Bilbao ("Why can't the "Bilbao Effect" be reproduced? | Bilbao as an example of Transformational Projects Action Planning") and Edmonton ("Downtown Edmonton cultural facilities development as an example of "Transformational Projects Action Planning""), and for infrastructure projects, the Purple Line light rail program in Suburban Maryland ("Codifying the complementary transit network improvements and planning initiatives recommended in the Purple Line writings") and the Silver Line Metrorail in Fairfax and Loudoun Counties in Virginia ("Using the Silver Line as the priming event, what would a transit network improvement program look like for Northern Virginia?").

At the scale of an individual project, the best example comes from my piece about best practice development of libraries, focused mostly on branch libraries, but not exclusively.

-- "Neighborhood libraries as nodes in a neighborhood and city-wide network of cultural assets," 2018

Flickr photo by Laura Billings of an Idea Store, with a farmers market out front.

One standout is how the Tower Hamlets borough of London merged their workforce development programs with the libraries, and relocated the libraries to well located places instead of in out of the way otherwise vacant buildings, using forward architecture and advanced branding and identity systems, to create "Idea Stores" ("The Idea Story," "Is this the library of the future?," BBC).

It's a particularly great example of how architecture is not enough.  They retained a now renowned architect, and he gets a lot of library commisions as a result.  But the other libraries afterward prove that it isn't design, but program, that makes for an innovative library.

Innovation Officers.  Along with the focus on big data, "Apps for Democracy," etc., a number of cities created "Chief Innovation Officers."  

I think that thrust has pretty much fallen by the wayside ("Whatever Happened to Innovation in Government?," Federal Computing Week) although Bloomberg Government News doesn't agree, "Chief innovation officers reconnect as their movement grows."

I was so impressed by the book Reinventing Government, back in the days of the Clinton Administration.  A promise that was never realized.